Inference on own data

[aksahveoffice]

I want to run the inference on our data using dales p retrained weights. my data classes is similar to dales data classes. how can i convert my data into proper format. or how can i run the inference on my data.

[drprojects]

Please have a look at the guidelines for creating your own dataset.

An important rule (that I will make clearer in future releases) for creating your own dataset is that your points must have labels within $[0, C]$, where:

• $C$ is your BaseDataset's num_classes
• giving the $C$ label to a point marks it as void/ignored/unlabeled (whichever you call it) and will exclude it from metrics and losses computation
• all labels $[0, C - 1]$ are assumed to be present in your dataset and will be used in metrics and losses computation

To this end, I recommend you make sure the output of your read_single_raw_cloud method never has labels outside of your $[0, C]$ range. Besides, if some labels in $[0, C - 1]$ are not useful to you (ie absent from your dataset), I recommend you densely remap your labels to another $[0, C_2 - 1]$ range (you can use torch_geometric.nn.pool.consecutive.consecutive_cluster for that, for instance), while making sure you only use the label $C_2$ for void/ignored/unlabeled points.